That could be cool! I know that google maps kind of already has that feature, where you can look at a few alternative routes and check walking distances. How could we factor in those learnings into the algorithms to maybe even encourage more walking? Thanks for the thoughts!
Tom, you're correct! The algorithms could be used to make dynamic routes, but I believe that'd be secondary. Step 1 would be making transfers easier to reduce travel time, and then building these better routes into the schedule.
I've had a similar experience to what Tiffany is describing. I live in East hills, so sometimes I opt to walk downtown, which is about 25 minutes away on foot, but at night or in bad weather I'd rather use a car or bus. I've found it to be usually cheaper for me to uber back and forth rather than look for parking. Uber is like ~6 dollars a ride, and the ramps are usually 12 dollars flat. Plus with Uber I can save my time.